IMAGE DETAILS

Silk velvet double-panel, probably Italy for the Ottoman market, late 16th-early 17th century. Louvre, MAO 932

| Click to enlarge

29 February 2008

Michael Franses reports: ‘Istanbul, Isfahan, Delhi: Three Capitals of Islamic Art’ at Istanbul’s Sakip Sabanci Museum (SSM) brings together 169 works of art made in the Islamic world between the 15th and 18th centuries, selected from the recently combined holdings of the Louvre and the Musée des Arts Décoratifs in Paris. They were chosen specifically to demonstrate the common Timurid heritage of the capitals of the Ottoman, Safavid and Mughal Empires.

With so many masterpieces it is hard to encapsulate the exhibition in a brief review, so I have concentrated on the carpets and textiles. The selection of carpets (and the catalogue entries) were the responsibility of Susan Day, scientific advisor at the Louvre, whose collection numbers some thirty carpets, to which must be added a further 140 carpets in the recent long-term loan of Islamic works of art (also including about four hundred textiles) from the neighbouring Musée des Arts Décoratifs (MADP). The works selected from the combined collections allow visitors to see and compare, side-by-side, outstanding works of art from the same workshop, or with similar patterns.

The Ottoman section comprises 58 items, including manuscripts, paintings, ceramics and jade. Among the carpets is a large section from a multiple-medallion Cairene Ottoman carpet, with border fragments attached. In the centre of this fragment is a yellow-ground crenellated ogival medallion with spade-like pendants. In each of the corners are quartered medallions on a light-blue ground, and elegant large leaves fill the field. This is one of the most beautiful Ottoman carpets known.¹

Two ‘first-period’ Ushak carpets with ogival medallions dating from the second half of the 15th century (and probably no later than 1500) are presented, both with stylised Kufic borders. One is the Louvre’s blue-ground carpet, acquired through Chevalier in 1995, with a field pattern of ‘autumn lotus’ leaves; the other is the Maciet red-ground carpet with a çintamani field of blue dots, from the MADP. Both are masterpieces. Also included are two well-known Ushak rugs from the Maciet gift at the MADP – a small Lotto arabesque rug with an open Kufic border and a keyhole niche rug with cloudband border.

Two Ottoman textiles from the Louvre stand out: one a double panel of Bursa velvet, acquired through Gérard Hadjer in 1994. Worked in brocaded metal thread that stands out against the deep red velvet ground, I have always suspected that the red on some earlier Bursa velvets could be from the insect dye lac, which might suggest a slightly earlier date, as cochineal was used extensively in Bursa from the third quarter of the 16th century. The silver thread is wrapped around a core of white or yellow silk to give the illusion of silver or gold. Precious metal thread was imported at great expense from Iran through Aleppo in such quantities that it threatened to bankrupt the Ottoman economy, and for a while the importation of gold thread was completely prohibited. The lattice with crowns was used in many centres in Italy and Spain during the 16th century. Other Bursa velvets with this pattern survive, including one in the V&A and one once with Edmond de Rothschild in Geneva, but none are as beautiful as the Louvre example. The pattern is most interesting: a cross-hatched ascending ogival lattice with large leaves and large crowns at the lattice junctions and a large tulip with a carnation above in each compartment. A charming feature is the presence of a number of narrow yellow lines in the velvet that run vertically through the length of the two panels (these also appears on another velvet, with a carnation design, and might well be indicative of a particular workshop). Almost the same design can be seen on what is arguably the ‘second-greatest’ Epirus embroidery, in the Art Institute, Chicago.²

However, arguably the very best surviving Epirus embroidery, the Alphonse Delort de Gleon coverlet with tulips from the Louvre is included in the SSM exhibition. This elegant textile, worked in silk on the very finest ‘linen’, is simply sublime, with an ogival lattice created by stems of ascending tulips, with single tulips in each of the offset diagonal compartments formed by the lattice. Two similar, much smaller, examples are known, both bohças, one in the Topkapi Saray, Istanbul,³ the other in a private collection in London.⁴

The rugs and textiles are placed close to cabinets of the very best 16th century Iznik ceramics depicting similar leaves and flowers. To give a taste of the colour of the Ottoman court, the exhibition would have been further enhanced with the addition of a few outstanding Istanbul seraser and Bursa lampas woven silks with gold thread.

The Safavid Empire is represented by 73 works of art: painted book-bindings, manuscripts, miniatures, ceramics, lacquerwork, one velvet and seven carpets. The latter are the very core of the two Paris collections. From the MADP come the controversial blue-ground vaq-vaq grotesque fragments, the Maciet half of the Paris-Cracow Paradise Park carpet, a section from a mid-16th century central Iranian hunting carpet, and a northwest Persian medallion carpet. From the Louvre we have the early 16th century Tabriz Paradise Park carpet from Mantes Cathedral; the 16th century Hakki Bey-Peytel small-silk Kashan with animals; and the Doistau late 16th century Kashan silk kilim. Just seven carpets, but a collection hard to top, as they are among the greatest masterpieces of Safavid art (Souren Melikian, please note).  

Susan Day has placed the MADP’s northwest Persian (Azerbaijan) medallion carpet at the beginning of the Iranian carpets in a small room alongside the çintamani-field medallion Ushak, demonstrating certain similarities in design. Several features on the Iranian carpet point to its great age (Day suggests the end of the 15th century): its well-proportioned, almost perfectly rounded central medallion; the exquisitely drawn cloud pattern around the edge of the medallion; and the long, elegant ivory clouds within the medallion. It is hard to find better. Slightly less satisfying is the very repetitive and rather unimaginative field of cloudbands and palmettes, both of which are well drawn in detail. Unusually, the outer and inner guard borders have the same design and proportions, albeit in different colours, while the balance between positive and negative space is well conceived, but they are rather simpler than the more intricate designs seen on other carpets of this genre. A number of red-wefted wool pile carpets such as this survive – some may well be early, while others can perhaps be dated to the end of the 16th century. There is also a strong and compelling case for several of the northwest Iranian carpets to be placed into the 15th century; but whether the Boston, Bardini and MADP carpets are among them, as suggested by Walter Denny, Roland Gilles, and Day, is yet to be proven.

The place and date of manufacture of the MADP’s blue-ground silk-foundation vaq-vaq fragments is also controversial. Day attributes them to Iran and the late 15th/early 16th century, as do Martin, Pope and, reportedly, Steven Cohen. Hanna Erdmann, on the other hand, and more recently Daniel Walker, have favoured an Indian attribution.⁵ I am not yet convinced either way.

The attribution of rugs is best based upon materials, structure and minor details. The fine wool that is used for the pile of the vaq-vaq fragments seems to be closer in feel to central Iranian carpets, such as the Ardabil medallion carpets and the Hatvany fragments,⁶ than to the few early wool pile Mughal carpets, such as the Ilchester and French & Co. examples,⁷ while the very finest carpets from both Iran and India have silk foundations. The primary reason for an Iranian attribution is comparison of the design with that of the Gulbenkian silk ‘tomb’ cover,⁸ which is certainly Iranian and from the beginning of the 16th century. However, although the animal grotesques and spiral vine are superficially similar, none of the flowers on the Paris fragments appear on the Gulbenkian rug, which has cloudbands, and the detailing of the stems and leaves is very different, as are the grotesque masks. The two carpets are probably from different places, but whether the Paris carpet comes from India or Iran must remain a question still to be resolved. Wherever and whenever it was made, its extraordinary beauty cannot be denied.⁹

Laid on a dais in the centre of the great double-height room at the SSM is the 15th/early 16th century blue-ground Tabriz ‘Paradise Park’ carpet originally from the Church of Notre Dame in Mantes. It was wonderful to be able to look down on this masterpiece from the upper level balcony, although I felt a little far away, yet down on the lower level, I felt a little too close to take in the boldness of the design. The wide border has huge palmettes alternating with paired phoenixes, the field has many animal combats, cypresses, peacocks and human figures with muskets, and the central medallion has dragons, phoenixes and, in the middle, leopards. The Mantes carpet does not have the quality of draughtsmanship and superb materials used in the Maciet ‘Paris-Cracow’ Paradise Park carpet, but it has a strength and clarity of design that can only be seen in the earliest Tabriz carpets, such as the great medallion fragment in the Musée des Gobelins collection.

The MADP’s glorious yellow-ground Paris-Cracow half-carpet is mounted on an incline to one side of the Mantes. Few people can remain unmoved by its sheer beauty, and the opportunity to see two ‘Paradise Park’ carpets side by side is rare and most appreciated. In materials, colours and weave it is certainly from central Iran. Day suggests a possible attribution to Herat, an opinion shared by Charles Grant Ellis and others. However, this would necessitate moving all structurally related carpets to Herat, as well as finding a new home for all those carpets currently attributed with far greater certainty to Khorasan, which have a very different weave, wool and palette – I don’t think so.

At the far end of the same room is the beautiful MADP red-ground silk foundation animal carpet fragment, probably from the middle of the 16th century, also tentatively attributed to Herat, and again more likely, to my mind, to be from central Iran. Between the MADP fragment and the Hakki Bey/Peytel small-silk Kashan with animals is a large late 16th century tile mosaic depicting a lion attacking a deer, echoing a theme seen on the carpets.

The small-silk Kashan is worth a visit on its own. Five large silk Kashan carpets survive, along with sixteen small rugs, four of which depict animals in landscapes.¹⁰ The small-silk Kashans are probably the most sensual and tactile of all oriental carpets. They have a quality of drawing that is not surpassed and a depth of colour that is truly magnificent. It is not simply the extraordinary colours or the animal combats that give this rug its great appeal. Much of its beauty is in the details: the drawing of the palmettes in the border, the elegance of the minor border, the fine tuning-fork design guard stripe. Exhibited at least fourteen times since the ‘Exposition d’art Musulman’ at the Palais de l’Industrie in 1893, and published almost forty times, it is undoubtedly the most famous of its type.

It is remarkably difficult to light large carpets and bring out the best of the objects on display in this high room. Similar problems were encountered at the 1983 Hayward Gallery exhibition ‘The Eastern Carpet in the Western World’, when some of the lending museums set strict restrictions on light levels. A partial solution was achieved by creating a darker setting and focused lights.

The Indian section has four carpets, three textiles and 34 other works of art. The carpets include the Jeuniette red-ground ‘grotesque’ fragments, made in Lahore in the first half of the 17th century, and the very fine ‘Kashmir’ lattice rug with shrubs made with pashmina pile on a silk foundation, also from the 17th century – both from the Louvre. From the MADP is the mid-17th century Tillot Lahore medallion carpet and a small floral rug with in-and out palmettes (I am not sure that the latter adds much to the exhibition). The textiles include a large 17th century velvet carpet with a lattice field, attached borders and silk fringes attached to the ends in the manner of classical Iranian carpets. There is a fragment with medallions and shrubs, also from the 17th century, and a sash with gold thread and shrubs, which can be attributed to India only with difficulty, because almost identical items are also attributed to Iran, as the catalogue points out.

This excellent exhibition’s success is the result of close collaboration between Dr Nazan Ölçer and Dr Filiz Cagman from the SSM, and Sophie Markariou and Charlotte Maury from the Louvre, who brought together expertise and experience from Turkey and France. While several of the choices made themselves, others were long debated, and the results are well worth the visit. Nor should we underestimate the organisational skills of the SSM’s Selmin Kangal and Irem Konukçu.

That ‘Three Capitals’ happened at all is due to a number of fortuitous circumstances: an agreement of mutual co-operation signed between the Louvre and Sabanci University in March 2007; the fact that the Louvre’s Islamic collection is in storage pending the completion, three years hence, of new purpose-built galleries; and the vision of the board of Turk Telekom, who sponsor some of the most significant educational and cultural events in Turkey.

The beauty and importance of art from the Islamic world are not widely known or understood in either the Orient or the Occident. The wonderful permanent galleries of Islamic art in great museums are generally not well attended: visiting them all too often one sees just a few lost souls and the occasional wandering expert. So we are delighted when major institutions such as the Louvre and the Musée des Arts Décoratifs offer special loan exhibitions devoted to this art. In a remarkably short time under Dr Ölçer’s direction, the SSM, best known for its collection of Islamic calligraphy, has presented a succession of outstanding exhibitions, from blockbuster shows on Picasso, Genghis Khan and Rodin, to smaller educational displays. The museum and its staff have nurtured an international following of enthusiastic visitors who keep returning for the inspiration and magic they present.

‘Istanbul, Isfahan, Delhi: 3 Capitals of Islamic Art. Masterpieces from the Louvre Collection’, is at the Sakip Sabanci Museum (SSM), Emirgan, Istanbul until 1 June 2008.

Notes

1. Other sections are in the Victoria & Albert Museum, London (HALI 4/1, 1981, pp. 43-4), and the Metropolitan Museum of Art, New York (Joseph V. McMullan, Islamic Carpets, New York 1965, pl.5), while fragments of the border are shared between the National Museum, Stockholm, the Textile Museum, Washington DC, and the Museum of Islamic Art in Cairo.

2. Margaret Gentles, Turkish and Greek Island Embroideries, Art Institute of Chicago, 1964, pl.18.

3. No.20, Melak Celal, Türk Islemeleri, Istanbul 1939.

4. Colnaghi, Imperial Ottoman Textiles, London 1980, no.27, and Cittone, Tesori Ottomani del XVI e XVII Secolo, Milan1980, no.21.

5. HALI 4/3, 1982, pp.224-5, 236; Daniel Walker, Flowers Underfoot: Indian Carpets of the Mughal Era, New York 1997, pp.72-3, 160, 167.

6. The Ardabil carpets, central Persia, dated 946 AH (1539–40 AD), wool pile on a silk foundation: (1) 530 x 1052cm, Victoria & Albert Museum, London, no.272-1893; (2) 400 x 719cm, Los Angeles County Museum of Art, no.53.50.2. The Hatvany ‘Paradise Park’ carpet fragments: (1) Budapest medallion and field section, 191 x 211 cm, whereabouts unknown, see Arthur Upham Pope, A Survey of Persian Art, London & New York 1938–39, pl.1141 = HALI 30, 1986, p.37; (2) The Bardini-Yerkes-Pratt fragments, Brooklyn Museum, New York, no.36.213, see HALI 88, 1996, p.107; (3) The Baron section. 44 x 63 cm, Musée Historiques des Tissus, Lyon, no.26.799, see HALI 33, 1987, p.40, pl.1 and note 4 (with structural analysis); (4) The Kelekian section, see Sotheby’s New York, 3-5 February 1977, lot 291.

7. Walker, op.cit.,: ‘French & Co.’, pp.48-9; ‘Ilchester’ pp. 52-3.

8. The Gulbenkian Hori and Inscription Small Silk Kashan (?) Tomb Cover (?), 93 x 236cm, silk pile on a silk foundation. Gulbenkian Museum, Lisbon, no. T.113, see HALI 114, 2001, p.75, fig.1.

9. Further border fragments from the Paris vaq-vaq carpet survive in the Kestner Museum, Hannover, and the Fine Arts Museum, Boston. It has been proposed that two wider border fragments (Burrell Collection, Glasgow; National Museum, Stockholm), formed parts of the main border of this carpet, but Day suggests that they are too wide and more likely to come from a carpet from the same workshop. There is unlikely to be a resolution to this question in the near future, as the analyses published to date are insufficient in detail. This wonderful group of fragments could engage me in a lengthy discussion that is not for presentation here, and certainly not until all the ‘wide’ borders are placed side by side with the field fragments and ‘narrow’ borders so they can be physically compared.

10. Small-silk Kashan rugs with animals: (1) The Princezza-Chappy-Altman, 178 x 238cm, Metropolitan Museum of Art, New York, no.14.40.721: (2) The Ludlow-Ford, 175 x 232cm, Detroit Institute of Art, no.25.23; (3) The Aynard-Rockefeller, 168 x 234 cm, Carpet Museum, Tehran; (4) The Hakki Bey-Peytel, 109 x 124 cm, Louvre, Paris, no.6741.